This script is meant for the stable diffusion webui -> https://github.com/AUTOMATIC1111/stable-diffusion-webui
Clone or download this repository then manually Put the script process_png_metadata.py in the stable-diffusion-webui\scripts folder, reload the webui to access the script in the txt2img section.
Drag and drop or upload image files to modify a prompt and override default settings to modify images.
This file is specifically meant for a batch process of image files with metadata applicable to the txt2img section of the webui. Allows user to easily use the hires fix upscale to get better images instead of img2img upscale.
overall features include:
- upload multiple files to process
- add/delete tags from a positive prompt
- override default generation settings with image metadata
source references this discussion -> AUTOMATIC1111/stable-diffusion-webui#4938
made this script through feature request issue #7462 -> AUTOMATIC1111/stable-diffusion-webui#7462